Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2011 Mar 15.
Published in final edited form as: Cancer Treat Res. 2004;118:47–81. doi: 10.1007/978-1-4419-9129-4_3

Animal Models of Bone Metastasis

Thomas J Rosol 1, Sarah H Tannehill-Gregg 1, Bruce E LeRoy 1, Stefanie Mandl 2,4, Christopher H Contag 2,3,4
PMCID: PMC3057671  NIHMSID: NIHMS183514  PMID: 15043188

Abstract

BACKGROUND

Animal models are important tools to investigate the pathogenesis and develop treatment strategies for bone metastases in humans. However, there are few spontaneous models of bone metastasis despite the fact that rodents (rats and mice) and other animals (dogs and cats) often spontaneously develop cancer. Therefore, most experimental models of bone metastasis in rodents require injection or implantation of neoplastic cells into orthotopic locations, bones, or the left ventricle of the heart.

METHODS

The current study reviews the natural incidence and clinical manifestation of bone metastases of mammary and prostate carcinoma in animals, as well as the experimental models developed in mice using animal and human-derived neoplasms.

RESULTS

Rats, mice, dogs, and cats often develop spontaneous mammary carcinoma, but bone metastases are rare. Intact and neutered dogs develop prostate carcinoma that is usually androgen independent and may be associated with regional bone invasion or distant bone metastasis. Normal dog prostate tissue induces new bone formation in vivo and can serve as a model of osteoblastic metastasis without concurrent bone destruction. Experimental models of osteolytic, osteoblastic, and mixed osteolytic/osteoblastic bone metastases include syngeneic rodent neoplasms or human xenografts implanted at orthotopic sites (e.g., breast or prostate glands) in immunodeficient mice, injection of cancer cells into the left ventricle of the heart, or direct injection into bones. New transgenic mouse models of cancer have a low incidence of spontaneous bone metastasis, but cell lines derived from these tumors can be selected in vivo for increased incidence of bone metastasis. It is essential to validate and correctly interpret the lesions in models of bone metastasis to accurately correlate the data from animal models to human disease. Animal models have provided support for the “seed and soil” hypothesis of bone metastasis. However, the roles of vascular patterns in the metaphyses of long bones and rapid bone turnover in young animals in the pathogenesis of metastasis in experimental models are uncertain. Improvements in the imaging of experimental animals in vivo using fluorescent markers or light emitted from luciferase have led to increased sensitivity of detection and more accurate quantification of bone metastases. For example, imaging of human prostate carcinoma PC-3M cells transfected with luciferase, following injection into the left ventricle, has demonstrated that there is rapid localization of tumor cells to bones and other organs, such as the kidneys and lungs.

CONCLUSIONS

Animal models of metastasis have supported drug development and have been useful for identification of metastasis suppressor and promoter genes as novel targets for the development of novel therapies. Further refinement of these models will involve spatiotemporal analysis of the metastatic process by imaging and use of image data to stage disease and guide tissue sampling for gene expression profiling via gene array technology. In the future, integrated analyses of these models will be needed to understand the complexities of this important disease process.

Keywords: bone metastasis, animal model, breast carcinoma, prostate carcinoma, dog, mouse, rat, imaging, osteoblastic metastasis

Ideal animal models of human cancers that metastasize to bone would reproduce the genetic and phenotypic changes that occur with human cancers. These include invasion, vascular spread to bone, and proliferation and survival in the bone marrow microenvironment with subsequent modifications of bone structure. In addition, such models would be reproducible and progress rapidly to permit timely investigations. Based on the pathogenesis of cancer in rodents and smaller mammals, this ideal may represent an unrealistic and impractical goal. However, animal models of bone metastasis that mimic selected aspects of human disease have been utilized and refinements to the models will continue to be developed.

Because spontaneous bone metastasis in animals is uncommon, most animal models of bone metastasis must be derived experimentally. This limitation has resulted in the development of specific models that represent unique stages of human bone metastasis. Thorough characterization of these animal models is required to permit their appropriate use as a representation of human disease. This review includes information on the rather infrequent spontaneous bone metastasis in animals with mammary and prostate carcinoma, the current uses of these and other animal models of bone metastasis, and recent developments that will better model human disease. There is a role for animal models in the study of bone metastasis, as well as a need for refinement of these models to advance our understanding of this important manifestation of oncogenesis.

Cancer progression with resulting bone metastases requires genetic changes that permit tissue invasion at the site of the primary tumor, entry into the vasculature, localization to bone, exit from the vasculature, survival and proliferation in the bone marrow microenvironment, and modification of bone structure and function.1,2 The genetic changes include metastasis-enhancing and suppressing genes, many of which are currently being identified and characterized using animal models. Bone metastasis-enhancing (and suppressing) genes are associated with multiple cellular processes that occur normally during mammalian development. These genes regulate cell shape and migration, interactions with extracellular matrix and stroma, angiogenesis, apoptosis, proliferation, and proteins that are usually associated with normal bone function (such as bone matrix proteins and hormones/cytokines that regulate bone cell activity). Although genes associated with bone metastasis can be identified readily by screening techniques (e.g., gene arrays), the validation and characterization of these genes will require sophisticated animal models that closely mirror the pathophysiology of bone metastasis in humans.

Because of the relatively artificial nature of animal models of bone metastasis, it is necessary to define what is considered a bone metastasis. End-stage lesions are readily identifiable and usually reveal tumor cell proliferation in bone that modifies bone structure. These would be comparable to clinically significant bone metastases in humans. Overt lesions can be identified by radiography or histopathology in animals. However, quantification of metastases with these insensitive techniques likely underestimates the actual number of bone metastases. For example, radiography will only detect severe lesions and will not measure all bone metastases. In addition, radiography will not detect bone metastases that fail to induce severe bone lysis or induce formation of mineralized matrix. In contrast, overestimation may result from newer highly sensitive techniques. For example, polymerase chain reaction (PCR) may detect tumor cells in bone that are arrested in blood vessels or quiescent cells in the bone marrow that may not develop into metastases.3,4 Therefore, morphologic assessment is necessary to confirm the incidence and nature of bone metastases in each animal model.

Animal models of bone metastasis include spontaneous tumors that arise in rodents or small mammals (e.g., dogs and cats), syngeneic transplantation of spontaneously occurring rodent cancers, chemical induction of cancers in selected strains of rats and mice, newly developed transgenic mouse models, and xenografts of human tumors or cell lines derived from human cancers into immunodeficient rodents (e.g., nude mice and rats, and severely compromised immunodeficient [SCID] mice; Fig. 1). Xenografted tissue or cells can be injected, subcutaneously into the left ventricle of the heart, or directly into the tibia or femur. In addition, xenografts can be implanted into orthotopic locations, such as the mammary fat pad, lungs, or prostate gland. Human bone can be implanted in immunodeficient rodents or bone ossicles can be induced in the subcutis to serve as a site for metastasis or implantation of tumor tissue to investigate the interactions of cancers with bone tissue in a nontraditional site.5,6 Animal models have been used successfully to select variants of cell or tumor lines that have an increased incidence of metastasis to bone. Selection pressure on the derived variants must be maintained to sustain the desired phenotype.

FIGURE 1.

FIGURE 1

Open in a new tab

Animal models of bone metastasis include spontaneous tumors in rodents and domestic animals, such as dogs and cats; spontaneous tumors in inbred strains of rodents that can be maintained as tumor lines in syngeneic hosts; chemically induced tumors in rodents; transgene-induced tumors in mice; and reconstitution models such as combinations of tumor cells with stromal cells, implantation of human bone in mice, or the formation of ectopic bone ossicles in the subcutis. Xenografts of human tumors and cell lines derived from human cancers can be implanted into immunodeficient mice (such as nude mice) in the subcutis, injected into the left ventricle of the heart (a reliable method for inducing bone metastases), directly injected into bones such as the tibia or femur, injected into the tail vein (induces lung metastases), or injected orthotopically (such as in the mammary gland, prostate gland, or the lungs).

SPONTANEOUS MAMMARY CARCINOMA IN ANIMALS

Rats and mice frequently develop benign and malignant mammary neoplasms with the incidence dependent on the strain. Unfortunately, these may not be good models for human disease. Most spontaneous mammary carcinomas in mice and rats do not metastasize and have mild local tissue invasion.7 There is a low incidence of spontaneous metastasis to regional lymph nodes or the lungs and bone metastasis is very rare. In addition, most adenocarcinomas in rodents rapidly lose their estrogen responsiveness and may not be good models of estrogen-responsive neoplasms in humans. Spontaneous development of mammary neoplasms in mice are due, in some cases, to a retrovirus, mouse mammary tumor virus (MMTV), but the role of retroviruses in the pathogenesis of human mammary carcinoma in uncertain. Proviral DNA, similar to MMTV, has been detected in a high percentage of human mammary carcinomas.8 For these reasons, other models have been developed in rodents.

Dogs frequently develop benign and malignant mammary neoplasia with an incidence similar to that observed in humans.9 Dogs develop hyperplasia, ductal carcinoma in situ, complex neoplasms with epithelial and myoepithelial components, and mixed neoplasms with cartilaginous and osseous differentiation of myoepithelial cells. Spontaneous mammary neoplasia in dogs has been imaged using indocyanin green as an optical contrast agent.10 Indocyanin green is a red fluorescent dye with emission that is ideal for imaging in mammals and it has been approved for human use. Approximately 50% of dog mammary carcinomas metastasize to regional lymph nodes and the lungs, but bone metastases are infrequent. Cats also have a high incidence of spontaneous mammary neoplasia. In cats, the neoplasms are typically invasive adenocarcinomas or ductal carcinomas that have a high incidence of recurrence after surgical removal and metastasis to regional lymph nodes and the lungs. Bone metastases and benign mammary tumors are also rare in cats. In addition, cats have a unique phenomenon of bone metastasis in which primary lung carcinomas can metastasize specifically to the bones of the digits.11 Yet, the first clinical sign is usually lameness, not the respiratory disease associated with the primary neoplasm.

SPONTANEOUS PROSTATE CARCINOMA IN ANIMALS

Animals have a very low incidence of prostate carcinoma compared with humans. Spontaneous prostate carcinoma occurs most commonly in dogs and is rare in rodents and other animals, including nonhuman primates. Some strains of rats have an increased incidence of prostate neoplasia. We describe mouse models of prostate carcinoma in this review.

There are important anatomic differences between the prostate glands of rodents and humans. Humans have a single gland with multiple regions, which include the transitional zone near the urethra, the central zone, the peripheral zone, and the anterior fibromuscular zone. Benign prostatic hyperplasia (BPH) is most common in the transitional zone and prostate carcinoma is most common in the peripheral zone. Rats and mice have four distinct lobes to the prostate gland. The anterior prostate (coagulating gland) extends rostrally along the ventral aspect of the seminal vesicle. The dorsal, lateral, and ventral prostate gland lobes extend around the base of the penis near the neck of the urinary bladder. The prostate gland in rodents is composed of compound ductules that lack the true acini that exist in humans and each lobe has a unique branching pattern.12 Rats and mice have a very low incidence of spontaneous proliferative lesions that develop in the prostate glands as they age.13,14 Secondary neoplasms, such as lymphoma, occur more commonly in aged rodents than do primary prostate neoplasms.

Specific strains of rats have an increased incidence of prostate neoplasms, including the Lobund Wistar and ACI/Seg rats. Up to 30% of aged (older than 20 months) Lobund Wistar rats develop prostate carcinoma in the anterior prostate/seminal vesicle complex.15 Administration of methynitrosourea (MNU) and testosterone increases the incidence and lowers the age at which cancer develops in these rats. Approximately 90% of rats will develop prostate carcinoma by 12 months of age.16 Lobund Wistar rats have high circulating concentrations of testosterone, which may predispose them to the development of prostate carcinoma. Initially, the carcinomas are testosterone dependent, but as they progress they become testosterone independent. The carcinomas eventually expand into the dorsolateral lobes of the prostate gland and metastasize to the lymph nodes and lungs. Development of the prostate carcinomas can be suppressed by diet restriction, testosterone ablation, dihydrotestosterone, diets with soy protein containing isoflavones, tamoxifen, and a vitamin D analog.17,18 Cell lines have been developed from the prostate carcinomas, such as PA-I, II, III, and IV. The PA-III cell line induces both osteoblastic and osteolytic bone lesions when the carcinoma is transplanted adjacent to the calvarium or scapula.19 The ACI/Seg rats develop a high incidence (80%) of microscopic prostate neoplasia in the ventral lobes at 36 months of age and a moderate incidence (16%) of macroscopically evident prostate carcinoma at the same age.20 In contrast, Copenhagen rats have a 10% incidence of microscopic prostate carcinoma and a less than 1% incidence of macroscopically evident prostate carcinomas at 36 months of age.21

Dogs have a single-lobed prostate gland similar to humans, but it does not have different anatomic regions. The prostatic urethra traverses the gland and is surrounded by ducts that quickly arborize into branched alveolar glands. All intact male dogs will develop simple (glandular) and complex (glandular and stromal) forms of BPH as they age.22 Both intact and castrated male dogs can develop prostate carcinoma, but dogs do not have the high incidence of prostate carcinoma that occurs in humans. The incidence of prostate carcinoma in castrated dogs is equal to or greater than intact dogs. Most cases of prostate carcinoma in dogs are androgen independent and expression of the testosterone receptor is uncommon. Initially, prostate carcinoma in dogs is confined to the gland and can result in dysuria, hematuria, or constipation. Many cancers will eventually invade the urinary bladder, pelvic cavity, lumbar vertebrae and pelvis, and some will metastasize to distant bones or organs. In some cases, the initial presenting feature is lameness due to a bone metastasis. The bone metastases are typically a mixture of osteoblastic and osteolytic changes, although the osteoblastic component can be the predominant pattern.

Prostate carcinomas in dogs have multiple morphologic patterns that include adenocarcinoma, intraalveolar carcinoma, papillary carcinoma, cribiform pattern, anaplastic carcinoma, and transitional cell-like carcinoma.22,23 Multiple patterns often exist in the same carcinoma. The cell of origin of prostate carcinoma in the dog has not been determined definitively, but immunohistochemical studies have led to the conclusion that the neoplasms arise from ductal cells.24 It is possible that the carcinomas arise from the ductal epithelium and then differentiate into ductal, glandular, and transitional cell patterns. Prostatic intraepithelial neoplasia (PIN) has been reported in dogs, but most cases occurred in dogs with overt carcinoma. Therefore, it has not been proven that PIN is a preneoplastic lesion in dogs.25 The incidence of PIN in aged, intact, and castrated dogs without carcinoma is unknown. Prostatic intraepithelial neoplasia occurred in 3% of aged military working dogs without prostate carcinoma and in 72% of the dogs (n = 25) that had prostate carcinoma.26

Prostate specific antigen (PSA) cannot be used as a marker for prostate carcinoma in dogs because they do not express this specific serine protease.27 The dog homolog to PSA is canine prostate specific arginine esterase (AE), which has 58% homology to PSA and has similar enzymatic activity as PSA on natural protein substrates. Prostate specific antigen has chymotrypsin-like activity and AE has trypsin-like activity on synthetic substrates. Canine AE is produced under androgen control by prostate glandular epithelial cells. Therefore, PSA and AE are related, but distinct, enzymes.

A canine prostate carcinoma cell line (DPC-1) has been transplanted successfully in the prostate of an aged dog that was immunosuppressed with cyclosporine.28 The prostate carcinoma grew in the normal prostate gland and invaded the pelvic cavity and regional iliac lymph nodes. Although bone metastases were not observed, the study demonstrated the potential for using dogs to investigate the pathogenesis and treatment of transplantable prostate carcinoma in a large animal model.

BONE INDUCTION BY NORMAL AND CANCEROUS CANINE PROSTATE TISSUE

When prostate carcinoma invades the prostate gland capsule and infiltrates the pelvic cavity in dogs, the neoplastic tissue induces marked new woven bone production from the periosteum of the lumbar vertebrae and pelvis (Fig. 2). Eventually, the tumor invades the medullary cavity of the bones. In addition, prostate carcinoma in dogs will metastasize directly to the medullary cavity of bones in the appendicular or axial skeleton. Intramedullary bone metastases are typically osteoblastic or a combination of osteoblastic/osteolytic lesions (Fig. 3). Normal dog prostate tissue is also capable of inducing new bone formation in vivo. Prostate gland tissue was implanted adjacent to the calvarium of nude mice and periosteal new bone proliferation was induced and increased the thickness of the calvarium by 70% in 2 weeks (Fig. 4).29 The bone formation was dependent on the presence of tissue containing both glands and stroma. Cultured epithelial or stromal cells alone did not induce the periosteal new bone formation. These data demonstrated that new bone formation induced by prostate carcinoma in vivo may represent an inherent characteristic of prostate tissue and may not be dependent on the cancer phenotype. It is likely that the new bone formation is due to growth factors secreted by the prostate epithelial and/or stromal cells and may be dependent on paracrine interactions between the two cell types.

FIGURE 2.

FIGURE 2

Open in a new tab

Prostate carcinoma in the dog (macroscopic photograph and radiograph of the caudal lumbar vertebrae and sacrum). The tumor invaded the pelvic cavity and induced marked proliferation of periosteal new bone formation along the ventral surfaces of the bodies of the lumbar vertebrae and sacrum. There also was invasion into the medullary cavity of the sacrum (best seen as an area of lysis on the left). Scale bar = 1 cm.

FIGURE 3.

FIGURE 3

Open in a new tab

Prostate carcinoma in the dog (macroscopic photograph and Faxitron radiograph of the distal femur). Two predominantly osteoblastic and partially osteolytic bone metastases of prostate carcinoma in the diaphysis (ca) and metaphysis/epiphysis. Notice the extensive proliferation of new woven bone trabeculae in the diaphysis and metaphysis induced by the prostate carcinoma (regions between the arrows). Scale bar = 1 cm.

FIGURE 4.

FIGURE 4

Open in a new tab

Mouse calvaria. (A) Control calvarium. (B) Calvarium demonstrating marked periosteal new bone proliferation on the convex surface induced by implantation of normal canine prostate in the subcutis (not shown) for 2 weeks.

SYNGENEIC MODELS OF BONE METASTASIS IN MICE, RATS, AND RABBITS

Many syngeneic models of mammary and prostate carcinoma in rats and mice do not metastasize readily to bone. However, sublines of the cancers can be selected in vivo that have an increased incidence of bone metastasis after orthotopic or intracardiac administration. For example, the 4T1.2 subclone of the 4T1 subline of a spontaneous mammary gland carcinoma from a Balb/cfC3H mouse has an increased incidence of metastasis to bone after injection into the mammary fat pad (orthotopic) or left ventricle of the heart.30 The MatLyLu androgen-insensitive subline of the rat Dunning prostate carcinoma (R3327) has been used as a reliable model for bone metastasis in vivo.31 Copenhagen rats develop hind limb paralysis 2–3 weeks after left ventricular injection of MatLyLu cells due to metastases in the lumbar vertebrae.32 The osteolytic bone metastases do not mimic the osteoblastic or mixed osteolytic/osteoblastic metastases of human prostate carcinoma. The PA-III cell line derived from a spontaneous prostate adenocarcinoma in a Lobund Wistar rat induced both osteolytic and osteoblastic reactions when transplanted adjacent to bone.19 This tumor line may be useful to investigate the ability of prostate carcinoma to induce new woven bone proliferation. The VX2 carcinoma was derived from a Shope papilloma virus-induced neoplasm in a domestic rabbit. Injection of VX2 cells into the tibia or ileum of rabbits induced mixed osteolytic and osteoblastic lesions.33

CHEMICAL INDUCTION OF MAMMARY AND PROSTATE CARCINOMA IN RATS AND MICE

Mammary neoplasia can be induced in rats by administering dimethylbenzanthracene, MNU, and N-ethyl-N-nitrosourea (ENU).34 Mammary adenocarcinomas induced by ENU in Sprague-Dawley rats may metastasize to the lungs. The rats often develop mild hypercalcemia, but bone metastases do not occur spontaneously.35,36 Prostate and seminal vesicle adenocarcinomas can be induced in Noble rats with testosterone/estradiol or MNU/testosterone combinations. An increased incidence of prostate adenocarcinomas can be induced in Lobund Wistar rats with MNU and testosterone.37 These tumors uncommonly metastasize to the lymph nodes and lungs and do not metastasize to bone.

TRANSGENIC INDUCTION OF MAMMARY AND PROSTATE CARCINOMA IN MICE

Oncogene expression can be targeted to the mammary and prostate glands using tissue selective promoters.38,39 The whey acidic protein, C(3)1, and MMTV promoters are often used for the mammary gland, whereas the probasin, C(3)1, and PSA promoters are used for the prostate gland. The advantages of transgenic models of cancer include their predictability and the autochthonous development of cancer (i.e., originating where normally found). A consensus report for the pathology of mammary carcinomas in genetically modified mice concluded that transgenes usually induced characteristic phenotypes and some of the neoplasms developed morphologic similarities to human mammary carcinoma.39 The disadvantage of the transgenic models is the low incidence of metastasis, especially bone metastases, often due to rapid progression of the primary neoplasm. Sublines of a spontaneous bone metastasis from a prostate carcinoma in a TRAMP mouse (transgenic mouse with the rat probasin promoter and expression of the SV40 early genes) have been developed that will permit investigations on bone metastasis in this transgenic model.40

HUMAN CANCER XENOGRAFTS: INTRACARDIAC INJECTION AND ORTHOTOPIC TRANSPLANTATION

Injection of human cancer cell lines into the left ventricle of nude mice induces bone metastases in vivo.41 This model has many advantages and has enabled testing of Paget’s “seed and soil” hypothesis with cancer cells serving as the seeds and the bone marrow microenvironment serving as the soil.42 Data supporting the seed and soil hypothesis include the selective growth of certain cell lines in the bone microenvironment to form bone metastases after intracardiac injection and the inhibition of tumor growth by inhibiting bone resorption.43,44

The disadvantage of the intracardiac injection model is the uncertain pathogenesis of bone metastasis after left ventricular injection of tumor cells. Many metastases occur in the metaphyses of the long bones, which are sites of active bone modeling and remodeling in young mice. Active bone turnover, high blood flow, and fenestrated sinusoids at these sites may predispose the metaphyses to the development of tumor growth. Blood vessels at the metaphyses have 180° turns and are common sites for embolization in young animals. Therefore, the anatomic arrangement of blood vessels in metaphyses may predispose to tumor cell embolization and development of bone metastases in growing rodents.45

Human cancer cells or tumor tissue can also be xenografted into orthotopic sites (e.g., the prostate or mammary glands, lungs, or long bones) to model bone metastasis in vivo. Orthotopic injection of cells into the prostate or mammary glands usually results in a low incidence of bone metastasis from late-stage cancers. Better results for modeling of bone metastasis may be obtained by using tumor tissue rather than cells. Tumor tissue can be produced artificially using human cell lines by implanting the cells subcutaneously in nude mice to form hybrid tumors with a mixture of human cancer cells and murine stroma and blood vessels. This technique has been used with hybrid mouse/human tissue using the human MDA-MB-435 cells, which resulted in bone metastases after implantation of tumor tissue in the mammary glands of nude mice.46 Human prostate carcinoma cells lines have been transplanted orthotopically in mice and utilized to successfully represent different stages of cancer progression in vivo ranging from androgen-dependent growth, androgen independence, androgen insensitivity, androgen repression, and metastatic behavior, including bone metastases.47 Few studies have investigated the metastasis of human cancers after intrapulmonary administration of tumor cells or tissue, but this technique may have utility as a model for human lung carcinoma.46 The mouse has a complete mediastinum, which permits intrathoracic surgery without positive-pressure ventilation of the lungs. Unilateral xenotransplantation of tumor tissue permits the growth of a large primary tumor and may allow enough time for metastasis to occur before tumor cachexia results. In addition, hybrid nude rat/human lung carcinoma tissue was implanted into the lungs of nude rats using an endotracheal cannula, which resulted in a 75% incidence of bone metastases.48

Few bone metastasis models mimic osteoblastic metastases. As a result, intraosseous injections have been used to model prostate carcinoma osteoblastic or mixed osteolytic/osteoblastic metastases.49 It is noteworthy that human mammary carcinoma cell lines (ZR-75-1 and MCF-7/Neu) have been used to model osteoblastic metastases in vivo and have demonstrated important roles of platelet-derived growth factor-BB and endothelin-1 in the pathogenesis of osteoblastic metastases.50,51 Typically, most mammary carcinomas induce osteolytic metastases in vivo.52 It is essential to assess accurately the pathology of animal models of bone metastasis, especially the models that are used to reproduce osteoblastic lesions. If a cancer results in severe lysis of cortical bone or induces pathologic fractures, then an intense proliferation of periosteal woven bone (Codman triangle) is expected. It is important not to interpret this reaction as an osteoblastic lesion.32 Osteoblastic metastases of human prostate carcinoma induce new woven bone on the surface of preexisting medullary trabecular bone. The most accurate models will reproduce this phenomenon. It is also possible that tumor induction of woven bone on the periosteum by cancer cells could mimic the pathogenesis of osteoblastic metastases if the bone proliferation is not secondary to disruption of the cortex.

SUBCUTANEOUS TRANSPLANTATION OF HUMAN BONE

Human fetal bone has been transplanted successfully to the subcutis of immunodeficient SCID mice to serve as a preferential site of metastasis of human prostate carcinoma cells injected into the left ventricle of the heart or as a site of tumor growth after local injection of cancer cells near the viable bone substrate.5 This model demonstrated the preferential selection of human cancer cells to localize and proliferate in human bone compared with mouse bones. Similar results were also reported with the use of adult human bone in SCID mice.53

IN VIVO IMAGING OF BONE METASTASES IN ANIMAL MODELS

The study of bone metastasis in small animal models has often relied on histologic analyses, PCR amplification, and radiography. However, assays that utilize excised tissues are subject to sampling limitations and cannot assess the overall extent of disease in a given animal subject.54–56 To overcome this limitation, in vivo detection of bone metastases using high-resolution Faxitron radiography has been used in attempt to provide temporal information. Radiography detects large osteolytic lesions. However, it does not detect micrometastatic lesions, the desired targets for therapeutic intervention.

Newer imaging modalities based on the optical detection of reporter genes promise to improve throughput and accuracy of quantitation in animal models of bone metastases.54,56,57 Transfection of tumor cells with genes encoding fluorescent or bioluminescent reporter protein has enabled in vivo imaging and longitudinal studies. In vivo fluorescence imaging and in vivo bioluminescent imaging (BLI) are inherently different techniques. For example, with fluorescence imaging, excitation light travels through the tissue to the tumor and emitted fluorescent light travels back to the detector. With BLI, the chemical substrate, luciferin, is distributed systemically and only the light emitted from the labeled cells travels from the tissue. Autofluorescence of tissue reduces the signal-to-noise ratio (SNR) in fluorescent imaging and the SNR is expected to be greater in BLI.58 Luciferase activity can be measured readily ex vivo using tissue extracts and fluorescence can be assayed by microscopy in tissue sections as well as in dissociated cells via flow cytometry. Therefore, combination reporter genes that are bioluminescent and fluorescent may offer the best of both approaches.59 Instrumentation for optical imaging employs imaging systems based on sensitive charge-coupled devices. These imaging systems are less expensive than those used in other imaging modalities (e.g., magnetic resonance imaging). These optical techniques typically require that the cells stably express reporter genes.

In vivo BLI has been used in a few mouse models of bone metastasis. In a xenograft model, small numbers of human mammary carcinoma cells were detectable in the bone marrow of immunodeficient mice by BLI.54 In this study, the BLI data were cross-validated using radiography at different stages of disease. The bioluminescent signal was detectable much earlier in the disease than the radiographic detection of osteolytic lesions. We have used BLI to study human prostate carcinoma cells (PC-3M-luc) in a xenograft model after intracardiac injection of 100,000 cells in nude mice (Fig. 5). Mice were imaged using the IVIS imaging system (Xenogen Corp., Alameda, CA) after injection of 3 µg of luciferin (Biosynth International, Naperville, IL). Within 3 minutes, cancer cells were homogenously distributed in all areas of the body (Fig. 5). Subsequently, at 15 minutes, cancer cells were localized to specific organs such as the lungs, kidneys, long bones, bones of the head, and eyes. It is noteworthy that some organs with a clearly detectable signal at these early time points, such as the kidneys, rarely develop overt metastases. At 24 hours, no viable tumor cells were visualized, demonstrating that most injected cells died (or were metabolically inactive) after intracardiac administration. On day 20, the first signals indicating bone metastases in the long bones and the spine were apparent. This occurred approximately 2 weeks before the detection of bone metastases by Faxitron radiography. In vivo imaging is a useful technique to determine the immediate success of intracardiac injection because some injections may result in tumor cell deposition into the pericardial or pleural cavities. Bone marrow micrometastases can elude radiographic detection, but these lesions were detected by BLI. Detection of micrometastatic lesions and the ability to follow tumor growth at multiple tissue sites over time will greatly enhance the study of metastatic disease.

FIGURE 5.

FIGURE 5

Open in a new tab

In vivo bioluminescent imaging of light emitted from luciferase in human PC-3M-luc prostate carcinoma cells in a nude mouse. Three minutes after injection of 100,000 cells into the left ventricle of the heart, cells were apparent throughout the body with very early localization in the kidney (light green focus). At 15 minutes, the signals from labeled tumor cells localized to the lungs, kidneys, bones (spine, long bones, and maxilla and mandible), and the eyes. There was no detectable signal at 24 hours (data not shown), indicating that most of the carcinoma cells died or were metabolically inactive. Initial metastases were evident in the long bones (blue focus) and spine on Day 20. The scale bar on the right is the relative intensity of light, which is an indirect measure of cell number.

SUMMARY AND CONCLUSIONS

Animal models will continue to be indispensable to investigate the pathogenesis of bone metastasis in vivo, to conduct preclinical chemotherapeutic, chemoprevention, and genetic therapy studies, to test gene delivery mechanisms, and to identify metastasis suppressor and inducer genes. It is likely that the bone marrow microenvironment (i.e., endothelial, stromal, hematopoietic, and bone cells) and the intercellular matrix play important roles in the localization and clonal growth of cancer cells in bone. Given the complexity of bone metastasis, many genes are expected to be involved in the pathogenesis and few are likely to be indispensable. The use of genomic and proteomic approaches to study these animal models will identify key targets for therapeutic intervention. As we further refine these models and use imaging for real-time evaluation of cells, and eventually target genes, these models will more closely mirror human disease and will hopefully become more predictive of the human response to therapy.

Acknowledgments

Supported by the United State Public Health Service, National Institutes of Health grant numbers CA77911 and RR00168 (T.J.R.) and CA88303 (C.H.C.).

Footnotes

Presented at the Third North American Symposium on Skeletal Complications of Malignancy, Bethesda, Maryland, April 25–27, 2002.

REFERENCES

  • 1.Yoneda T, Williams PJ, Myoi A, Michigami T, Mbalaviele G. Cellular and molecular mechanisms of development of skeletal metastases. In: Body J-J, editor. Tumor bone diseases and osteoporosis in cancer patients. New York: Marcel Dekker; 1999. pp. 41–69. [Google Scholar]
  • 2.Cher ML. Mechanisms governing bone metastasis in prostate cancer. Curr Opin Urol. 2001;11:483–488. doi: 10.1097/00042307-200109000-00006. [DOI] [PubMed] [Google Scholar]
  • 3.van der Pluijm G, Sijmons B, Vloedgraven H, Deckers M, Papapoulos S, Lowik C. Monitoring metastatic behavior of human tumor cells in mice with species-specific polymerase chain reaction: elevated expression of angiogenesis and bone resorption stimulators by breast cancer in bone metastases. J Bone Miner Res. 2001;16:1077–1091. doi: 10.1359/jbmr.2001.16.6.1077. [DOI] [PubMed] [Google Scholar]
  • 4.Sung V, Cattell DA, Bueno JM, et al. Human breast cancer cell metastasis to long bone and soft organs of nude mice: a quantitative assay. Clin Exp Metastasis. 1997;15:173–183. doi: 10.1023/a:1018409028662. [DOI] [PubMed] [Google Scholar]
  • 5.Nemeth JA, Harb JF, Barroso U, Jr, He Z, Grignon DJ, Cher ML. Severe combined immunodeficient-hu model of human prostate cancer metastasis to human bone. Cancer Res. 1999;59:1987–1993. [PubMed] [Google Scholar]
  • 6.Schneider A, Taboas JM, Krebsbach PH, McCauley LK. An ectopic tissue engineered bone model to study hormonal responses in vivo [abstract]. Presented at the Third North American Symposium on Skeletal Complications of Malignancy; April 25–27, 2002; Bethesda, MD. [Google Scholar]
  • 7.Seely JC, Boorman GA. Mammary gland and specialized sebaceous glands. In: Maronpot RR, Boorman GA, Gaul BW, editors. Pathology of the mouse. Vienna, IL: Cache River Press; 1999. pp. 613–636. [Google Scholar]
  • 8.Liu B, Wang Y, Melana SM, et al. Identification of a proviral structure in human breast cancer. Cancer Res. 2001;61:1754–1759. [PubMed] [Google Scholar]
  • 9.Misdorp W. Tumors of the mammary gland. In: Meuten DJ, editor. Tumors in domestic animals. 4th ed. Ames: Iowa State Press; 2002. pp. 575–606. [Google Scholar]
  • 10.Hawrysz DJ, Sevick-Muraca EM. Developments toward diagnostic breast cancer imaging using near-infrared optical measurements and fluorescent contrast agents. Neoplasia. 2000;2:388–417. doi: 10.1038/sj.neo.7900118. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.van der Linde-Sipman JS, van den Ingh TS. Primary and metastatic carcinomas in the digits of cats. Vet Q. 2000;22:141–145. doi: 10.1080/01652176.2000.9695043. [DOI] [PubMed] [Google Scholar]
  • 12.Cunha GR, Donjacour AA, Cooke PS, et al. The endocrinology and developmental biology of the prostate. Endocr Rev. 1987;8:338–362. doi: 10.1210/edrv-8-3-338. [DOI] [PubMed] [Google Scholar]
  • 13.Suwa T, Nyska A, Haseman JK, Mahler JF, Maronpot RR. Spontaneous lesions in control B6C3F1 mice and recommended sectioning of male accessory sex organs. Toxicol Pathol. 2002;30:228–234. doi: 10.1080/019262302753559560. [DOI] [PubMed] [Google Scholar]
  • 14.Suwa T, Nyska A, Peckham JC, et al. A retrospective analysis of background lesions and tissue accountability for male accessory sex organs in Fischer-344 rats. Toxicol Pathol. 2001;29:467–478. doi: 10.1080/01926230152500086. [DOI] [PubMed] [Google Scholar]
  • 15.Pollard M. Lobund-Wistar rat model of prostate cancer in man. Prostate. 1998;37:1–4. doi: 10.1002/(sici)1097-0045(19980915)37:1<1::aid-pros1>3.0.co;2-l. [DOI] [PubMed] [Google Scholar]
  • 16.Pollard M, Luckert PH, Snyder DL. The promotional effect of testosterone on induction of prostate-cancer in MNU-sensitized L-W rats. Cancer Lett. 1989;45:209–212. doi: 10.1016/0304-3835(89)90079-7. [DOI] [PubMed] [Google Scholar]
  • 17.Pollard M. Prevention of prostate-related cancers in Lobund-Wistar rats. Prostate. 1999;39:305–309. doi: 10.1002/(sici)1097-0045(19990601)39:4<305::aid-pros12>3.0.co;2-y. [DOI] [PubMed] [Google Scholar]
  • 18.Pollard M. Dihydrotestosterone prevents spontaneous adenocarcinomas in the prostate-seminal vesicle in aging L-W rats. Prostate. 1998;36:168–171. doi: 10.1002/(sici)1097-0045(19980801)36:3<168::aid-pros4>3.0.co;2-d. [DOI] [PubMed] [Google Scholar]
  • 19.Koutsilieris M. PA-III rat prostate adenocarcinoma cells [review] In Vivo. 1992;6:199–203. [PubMed] [Google Scholar]
  • 20.Varma VA, Austin GE. Morphologic characterization of early prostatic carcinomas in the ACI rat: a light and electron microscopic study. Exp Mol Pathol. 1990;52:202–211. doi: 10.1016/0014-4800(90)90005-x. [DOI] [PubMed] [Google Scholar]
  • 21.Isaacs JT. The aging ACI/Seg versus Copenhagen male rat as a model system for the study of prostatic carcinogenesis. Cancer Res. 1984;44:5785–5596. [PubMed] [Google Scholar]
  • 22.Leav I, Schelling SH, Merk FB. Age-related and sex hormone-induced changes in the canine prostate. In: Mohr U, Carlton WW, Dungworth DL, Benjamin SA, Capen CC, Hahn FA, editors. Pathobiology of the aging dog. Ames: Iowa State Press; 2001. pp. 310–329. [Google Scholar]
  • 23.Cornell KK, Bostwick DG, Cooley DM, et al. Clinical and pathologic aspects of spontaneous canine prostate carcinoma: a retrospective analysis of 76 cases. Prostate. 2000;45:173–183. doi: 10.1002/1097-0045(20001001)45:2<173::aid-pros12>3.0.co;2-r. [DOI] [PubMed] [Google Scholar]
  • 24.Leav I, Schelling KH, Adams JY, Merk FB, Alroy J. Role of canine basal cells in postnatal prostatic development, induction of hyperplasia, and sex hormone-stimulated growth; and the ductal origin of carcinoma. Prostate. 2001;48:210–224. doi: 10.1002/pros.1100. [DOI] [PubMed] [Google Scholar]
  • 25.Waters DJ, Hayden DW, Bell FW, Klausner JS, Qian J, Bostwick DG. Prostatic intraepithelial neoplasia in dogs with spontaneous prostate cancer. Prostate. 1997;30:92–97. doi: 10.1002/(sici)1097-0045(19970201)30:2<92::aid-pros4>3.0.co;2-m. [DOI] [PubMed] [Google Scholar]
  • 26.Aquilina JW, McKinney L, Pacelli A, et al. High grade prostatic intraepithelial neoplasia in military working dogs with and without prostate cancer. Prostate. 1998;36:189–193. doi: 10.1002/(sici)1097-0045(19980801)36:3<189::aid-pros7>3.0.co;2-c. [DOI] [PubMed] [Google Scholar]
  • 27.Clements JA. The glandular kallikrein family of enzymes: tissue-specific expression and hormonal regulation. Endocr Rev. 1989;10:393–419. doi: 10.1210/edrv-10-4-393. [DOI] [PubMed] [Google Scholar]
  • 28.Anidjar M, Villette JM, Devauchelle P, et al. In vivo model mimicking natural history of dog prostate cancer using DPC-1, a new canine prostate carcinoma cell line. Prostate. 2001;46:2–10. doi: 10.1002/1097-0045(200101)46:1<2::aid-pros1002>3.0.co;2-5. [DOI] [PubMed] [Google Scholar]
  • 29.LeRoy BE, Bahnson RR, Rosol TJ. Canine prostate induces new bone formation in mouse calvaria: a model of osteoinduction by prostate tissue. Prostate. 2002;50:104–111. doi: 10.1002/pros.10037. [DOI] [PubMed] [Google Scholar]
  • 30.Lelekakis M, Moseley JM, Martin TJ, et al. A novel orthotopic model of breast cancer metastasis to bone. Clin Exp Metastasis. 1999;17:163–170. doi: 10.1023/a:1006689719505. [DOI] [PubMed] [Google Scholar]
  • 31.Tennant TR, Kim H, Sokoloff M, Rinker S. The Dunning model. Prostate. 2000;43:295–302. doi: 10.1002/1097-0045(20000601)43:4<295::aid-pros9>3.0.co;2-w. [DOI] [PubMed] [Google Scholar]
  • 32.Blomme EAG, Dougherty KM, Pienta KJ, Capen CC, Rosol TJ, McCauley LK. Skeletal metastasis of prostate adenocarcinoma in rats: morphometric analysis and role of parathyroid hormone-related protein. Prostate. 1999;39:187–197. doi: 10.1002/(sici)1097-0045(19990515)39:3<187::aid-pros7>3.0.co;2-3. [DOI] [PubMed] [Google Scholar]
  • 33.Galasko CS. Mechanisms of lytic and blastic metastatic disease of bone. Clin Orthop. 1982;169:20–27. [PubMed] [Google Scholar]
  • 34.Ip C. Mammary tumorigenesis and chemoprevention studies in carcinogen-treated rats. J Mammary Gland Biol Neoplasia. 1996;1:37–47. doi: 10.1007/BF02096301. [DOI] [PubMed] [Google Scholar]
  • 35.Stoica G, Koestner A, Capen CC. Characterization of N-ethyl-N-nitrosourea-induced mammary tumors in the rat. Am J Pathol. 1983;110:161–169. [PMC free article] [PubMed] [Google Scholar]
  • 36.Stoica G, Koestner A, Capen CC. Neoplasms induced with high single doses of N-ethyl-N-nitrosourea in 30-day-old Sprague-Dawley rats, with special emphasis on mammary neoplasia. Anticancer Res. 1984;4:5–12. [PubMed] [Google Scholar]
  • 37.Pollard M, Wolter WR, Sun L. Prostate-seminal vesicle cancers induced in noble rats. Prostate. 2000;43:71–74. doi: 10.1002/(sici)1097-0045(20000401)43:1<71::aid-pros10>3.0.co;2-h. [DOI] [PubMed] [Google Scholar]
  • 38.Huss WJ, Maddison LA, Greenberg NM. Autochthonous mouse models for prostate cancer: past, present and future. Semin Cancer Biol. 2001;11:245–260. doi: 10.1006/scbi.2001.0373. [DOI] [PubMed] [Google Scholar]
  • 39.Cardiff RD, Anver MR, Gusterson BA, et al. The mammary pathology of genetically engineered mice: the consensus report and recommendations from the Annapolis meeting. Oncogene. 2000;19:968–988. doi: 10.1038/sj.onc.1203277. [DOI] [PubMed] [Google Scholar]
  • 40.Foster BA, Greenberg NM. New model of bone metastasis for prostate cancer: cell lines derived from a bone metastasis in TRAMP [abstract] Proc Am Assoc Cancer Res. 2001;42:140. [Google Scholar]
  • 41.Yoneda T. Cellular and molecular basis of preferential metastasis of breast cancer to bone. J Orthop Sci. 2000;5:75–81. doi: 10.1007/s007760050012. [DOI] [PubMed] [Google Scholar]
  • 42.Paget S. The distribution of secondary growths in cancer of the breast. Lancet. 1889;1:571–573. [PubMed] [Google Scholar]
  • 43.Yoneda T, Michigami T, Yi B, Williams PJ, Niewolna M, Hiraga T. Use of bisphosphonates for the treatment of bone metastasis in experimental animal models. Cancer Treat Rev. 1999;25:293–299. doi: 10.1053/ctrv.1999.0133. [DOI] [PubMed] [Google Scholar]
  • 44.Zhang J, Dai J, Qi Y, et al. Osteoprotegerin inhibits prostate cancer-induced osteoclastogenesis and prevents prostate tumor growth in the bone. J Clin Invest. 2001;107:1235–1244. doi: 10.1172/JCI11685. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Yoneda T. Arterial microvascularization and breast cancer colonization in bone. Histol Histopathol. 1997;12:1145–1149. [PubMed] [Google Scholar]
  • 46.Hoffman RM. Orthotopic metastatic mouse models for anticancer drug discovery and evaluation: a bridge to the clinic. Invest New Drugs. 1999;17:343–359. doi: 10.1023/a:1006326203858. [DOI] [PubMed] [Google Scholar]
  • 47.Zhau HE, Li CL, Chung LW. Establishment of human prostate carcinoma skeletal metastasis models. Cancer. 2000;88:2995–3001. doi: 10.1002/1097-0142(20000615)88:12+<2995::aid-cncr15>3.3.co;2-p. [DOI] [PubMed] [Google Scholar]
  • 48.Howard RB, Mullen JB, Pagura ME, Johnston MR. Characterization of a highly metastatic, orthotopic lung cancer model in the nude rat. Clin Exp Metastasis. 1999;17:157–162. doi: 10.1023/a:1006637712294. [DOI] [PubMed] [Google Scholar]
  • 49.Corey E, Quinn JE, Bladou F, et al. Establishment and characterization of osseous prostate cancer models: intra-tibial injection of human prostate cancer cells. Prostate. 2002;52:20–33. doi: 10.1002/pros.10091. [DOI] [PubMed] [Google Scholar]
  • 50.Yi B, Williams PJ, Niewolna M, Wang Y, Yoneda T. Tumor-derived platelet-derived growth factor-BB plays a critical role in osteosclerotic bone metastasis in an animal model of human breast cancer. Cancer Res. 2002;62:917–923. [PubMed] [Google Scholar]
  • 51.Mohammad KS, Yin JJ, Grubbs BG, Cui Y, Padley R, Guise TA. Endothelin-1 (ET-1) mediates pathological but not normal bone remodeling [abstract] J Bone Miner Res. 2001;16 suppl 1:S453. [Google Scholar]
  • 52.Chirgwin JM, Guise TA. Molecular mechanisms of tumor-bone interactions in osteolytic metastases. Crit Rev Eukaryot Gene Expr. 2000;10:159–178. [PubMed] [Google Scholar]
  • 53.Yonou H, Yokose T, Kamijo T, et al. Establishment of a novel species- and tissue-specific metastasis model of human prostate cancer in humanized non-obese diabetic/severe combined immunodeficient mice engrafted with human adult lung and bone. Cancer Res. 2001;61:2177–2182. [PubMed] [Google Scholar]
  • 54.Wetterwald A, van der PG, Que I, et al. Optical imaging of cancer metastasis to bone marrow: a mouse model of minimal residual disease. Am J Pathol. 2002;160:1143–1153. doi: 10.1016/S0002-9440(10)64934-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.Sweeney TJ, Mailander V, Tucker AA, et al. Visualizing the kinetics of tumor-cell clearance in living animals. Proc Natl Acad Sci US A. 1999;96:12044–12049. doi: 10.1073/pnas.96.21.12044. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Edinger M, Sweeney TJ, Tucker AA, Olomu AB, Negrin RS, Contag CH. Noninvasive assessment of tumor cell proliferation in animal models. Neoplasia. 1999;1:303–310. doi: 10.1038/sj.neo.7900048. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57.Yang M, Baranov E, Moossa AR, Penman S, Hoffman RM. Visualizing gene expression by whole-body fluorescence imaging. Proc Natl Acad Sci U S A. 2000;97:12278–12282. doi: 10.1073/pnas.97.22.12278. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Rice BW, Cable MD, Nelson MB. In vivo imaging of light-emitting probes. J Biomed Opt. 2001;6:432–440. doi: 10.1117/1.1413210. [DOI] [PubMed] [Google Scholar]
  • 59.Costa GL, Sandora MR, Nakajima A, et al. Adoptive immunotherapy of experimental autoimmune encephalomyelitis via T cell delivery of the IL-12 p40 subunit. J Immunol. 2001;167:2379–2387. doi: 10.4049/jimmunol.167.4.2379. [DOI] [PubMed] [Google Scholar]

RESOURCES